Parallel
Computing Toolbox™ supports distributed arrays to
partition large arrays across multiple MATLAB® workers.
Simultaneous execution is supported by the single program multiple
data (spmd) language construct to facilitate
communication between workers. You can use distributed arrays in Parallel
Computing Toolbox to
run big data applications using the combined memory of your cluster.
distributed | Create distributed array from data in client workspace |
gather | Transfer distributed array or gpuArray to local workspace |
spmd | Execute code in parallel on workers of parallel pool |
Composite | Create Composite object |
parallel.pool.Constant | Build parallel.pool.Constant from data or function handle |
codistributed | Create codistributed array from replicated local data |
parpool | Create parallel pool on cluster |
delete (Pool) | Shut down parallel pool |
redistribute | Redistribute codistributed array with another distribution scheme |
codistributed.build | Create codistributed array from distributed data |
for | for-loop over distributed range |
getLocalPart | Local portion of codistributed array |
globalIndices | Global indices for local part of codistributed array |
gop | Global operation across all workers |
write | Write distributed data to an output location |
distributed | Access elements of distributed arrays from client |
codistributed | Access elements of arrays distributed among workers in parallel pool |
Composite | Access nondistributed variables on multiple workers from client |
codistributor1d | 1-D distribution scheme for codistributed array |
codistributor2dbc | 2-D block-cyclic distribution scheme for codistributed array |
parallel.Pool | Access parallel pool |
Use datastore or distributed to
create distributed arrays and partition the data among your workers
Using MATLAB Functions on Distributed Arrays
MATLAB functions that operate on codistributed arrays
Run Single Programs on Multiple Data Sets
Use spmd statements to run the
same code on multiple datasets and control codistributed arrays
Access Worker Variables with Composites
Composite objects in the MATLAB client session let you directly access data values on the workers.
Learn about starting and stopping parallel pools, pool size, and cluster selection.
Specify Your Parallel Preferences
Specify your preferences, and automatically create a parallel pool.
Nondistributed Versus Distributed Arrays
Describes the various types of arrays used in communicating jobs, including pmode
Working with Codistributed Arrays
Describes how to use codistributed arrays for calculation
Looping Over a Distributed Range (for-drange)
Describes how to program a for-loop
with codistributed arrays